Token导航 LogoToken导航TokenDH.com
AI 工具只读github未标认证来源可访问clear审计通过

team-templates团队模板

Agent Skill

team-templates 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

196

周安装

8

GitHub Stars

8

下载量

63
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:team-templates(团队模板)
来源仓库:https://github.com/hubert-dudek/medium
仓库路径:skills/team-templates
安装命令:
npx skills add https://github.com/hubert-dudek/medium --skill team-templates
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/hubert-dudek/medium --skill team-templates

简介

team-templates 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装,需结合原始 README 核验具体用法。
  • 安装前建议确认权限范围、维护状态及是否会触发联网、命令执行或文件读写操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Team templates library

Use this skill to generate consistent, copy/paste-ready templates for common data/analytics team workflows.

Template catalog

Pick the best match based on the user request:

  • Project kickoff / 1‑pagerassets/kickoff.md
  • Weekly status update (Slack + longer-form) → assets/weekly-status-update.md
  • Incident postmortem / RCA (Sev/Outage) → assets/incident-postmortem.md
  • Data contract (dataset/table contract in YAML) → assets/data-contract.yml
  • Unity Catalog access request (permissions + optional GRANT SQL) → assets/unity-catalog-access-request.md

How to use this skill

1) Identify the template type

If the user explicitly names one (kickoff, postmortem, weekly update, data contract, access request), use it.

If not explicit, infer from keywords:

  • *“RCA”, “postmortem”, “outage”, “incident”, “Sev”* → incident postmortem
  • *“1‑pager”, “kickoff”, “proposal”, “PRD”, “new project”* → kickoff doc
  • *“data contract”, “schema contract”, “dataset contract”* → data contract YAML
  • *“grant”, “permission”, “Unity Catalog”, “access to schema/table”* → UC access request
  • *“weekly update”, “status”, “what did we do this week”* → weekly status update

2) Ask for only missing inputs (max 3 questions)

If required details are missing, ask up to three targeted questions. Prefer defaults over interrogations.

Common inputs:

  • Names: project/dataset/table, team, owner
  • Dates: start date, incident date/time range
  • Audience: internal team vs stakeholders vs exec summary
  • Access scope: catalog.schema.table, permission level (SELECT/MODIFY/OWN), duration

If the user provided none, assume:

  • Audience: “internal + stakeholders”
  • Timezone: user local (unknown) → label as “local time”
  • Dates: today’s date is fine only if user asked “today”; otherwise leave placeholders.

3) Generate a filled template (deliverable-first)

Output the chosen template in Markdown (or YAML for the data contract), with placeholders replaced when you have information.

Rules:

  • Keep headings intact so the doc is scannable.
  • Keep checklists as - [] so teams can track work.
  • Use neutral language. Avoid blaming individuals.

4) Add a short “Next steps” section

After the template, add 3–8 concrete next steps tailored to the template type.

Examples

Example A — kickoff

User: “Create a kickoff doc for a new customer churn dashboard in Databricks.”

Assistant output: A filled version of assets/kickoff.md including goals, stakeholders, datasets, milestones, and a definition of done.

Example B — postmortem

User: “We had a 2‑hour outage because the DLT pipeline failed. Draft the postmortem.”

Assistant output: A filled version of assets/incident-postmortem.md with impact, timeline, root cause hypotheses, and action items.

Edge cases

  • User wants Confluence/Google Doc formatting: Output Markdown, but keep headings and checklists; it will paste cleanly.
  • User wants multiple templates: Output one at a time; ask which is highest priority.
  • User provides sensitive data: Keep it high-level; replace secrets with [REDACTED].
  • Unknown incident details: Keep placeholders and add a section “Information needed” listing what’s missing.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

27.95%
按下载量换算18

windsurf

23.92%
按下载量换算15

trae

16.62%
按下载量换算10

OpenCode

12.71%
按下载量换算8

Codex

6.8%
按下载量换算4

Antigravity

3.6%
按下载量换算2

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

来源信息

继续浏览同类 Skills